elite-copywriting
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns such as prompt injection, data exfiltration, persistence mechanisms, or obfuscation were identified in the skill.
- [SAFE]: The skill references the use of established third-party services and tools, such as PostHog (@posthog/wizard) and Vercel, within a legitimate professional context for deployment and conversion analytics. References to these well-known services are considered safe per the analysis guidelines.
- [SAFE]: The skill ingests untrusted external data (e.g., customer reviews, Reddit threads) as part of its audience research phase (Voice of Customer mining). This creates an inherent surface for indirect prompt injection; however, the skill mitigates this risk through a rigorous 8-phase workflow and prompts that emphasize verbatim extraction and structured synthesis rather than open-ended interpretation of the data.
Audit Metadata